Name: DuelLab GameBench 2 public release gb2-20260726-11fb0891c89b
Creator: DuelLab
License: https://creativecommons.org/licenses/by/4.0/

Game profile

Chess

Chess is an 8x8 tactical piece-movement game with asymmetric pieces, castling, en passant, promotion, check, checkmate, and automatic draw rules.

Scores here compare generated programs on DuelLab's encoded rules for Chess; they are not a broad measure of human play strength.

Entrants

Every admitted unique code identity for GB2-02: Chess. Elo and W/L/D use the public historical sequential-Elo result universe. Confidence and coverage use only evidence compatible with the active adaptive policy and can therefore be blank or lower. Capped draws are rated; unilateral model faults are listed separately.

Search model Status Sort Showing 40 entrants.

Entrant	Model	Status	Score	Elo	Confidence	W / L / D	Rated / attempted	Opponents	Faults / plugins	Generation
E-16462f3d1f3a7b19	Claude Opus 4.5	Under-covered	63.8	1500	—	0 / 0 / 6	6 / 6	3	0 / 1	First pass · 2026-07-02
E-c142942f4138c871	Claude Opus 4.8	Under-covered	63.8	1500	—	0 / 0 / 6	6 / 6	3	0 / 1	First pass · 2026-07-02
E-51ddad48e10b006f	Claude Opus 5	Under-covered	0.0	1500	—	0 / 0 / 2	2 / 2	1	0 / 1	First pass · 2026-07-25
E-772d1a710082ddb3	Claude Opus 5	Under-covered	54.1	1516	—	1 / 0 / 3	4 / 4	2	0 / 1	First pass · 2026-07-25
E-3358fbcda6ab6f4d	DeepSeek V4 Flash	Under-covered	78.1	1500	—	0 / 0 / 8	8 / 8	4	0 / 1	First pass · 2026-07-02
E-bc9a8c3886a48505	DeepSeek V4 Flash	Under-covered	89.3	1501	—	0 / 0 / 10	10 / 10	5	0 / 1	First pass · 2026-07-02
E-6c3b019c1b20a09b	DeepSeek V4 Pro	Under-covered	100.0	1515	—	1 / 0 / 9	10 / 10	5	0 / 1	First pass · 2026-07-02
E-1010911d64a09b68	Gemini 3.5 Flash Lite	Pending	—	—	—	0 / 0 / 0	0 / 0	0	0 / 1	Repaired · 2026-07-24
E-e65fdcb71f09a611	Gemini 3.5 Flash Lite	Pending	—	—	—	0 / 0 / 0	0 / 0	0	0 / 1	Repaired · 2026-07-24
E-4492f6d65cd6280b	Gemini 3.6 Flash	Pending	—	—	—	0 / 0 / 0	0 / 0	0	0 / 1	First pass · 2026-07-24
E-9c8b43dc41bad760	Gemini 3.6 Flash	Pending	—	—	—	0 / 0 / 0	0 / 0	0	0 / 1	Repaired · 2026-07-24
E-98ddce22d3eb5886	GPT-5.4	Under-covered	79.1	1501	—	0 / 0 / 8	8 / 8	4	0 / 1	First pass · 2026-07-02
E-322cbb8be1559087	GPT-5.4 Mini	Under-covered	88.3	1500	—	0 / 0 / 10	10 / 10	5	0 / 1	First pass · 2026-07-02
E-b45c06f99639b8db	GPT-5.4 Mini	Under-covered	86.7	1498	—	0 / 0 / 10	10 / 10	5	0 / 1	First pass · 2026-07-02
E-dd45dfa7984ab9fa	GPT-5.4 Mini	Under-covered	79.0	1501	—	0 / 0 / 8	8 / 8	4	0 / 1	Repaired · 2026-07-01
E-038a95fb13c8aa94	GPT-5.4 Nano	Under-covered	79.5	1489	—	0 / 1 / 9	10 / 10	5	0 / 1	Repaired · 2026-07-09
E-23b12c36ef5c7ef8	GPT-5.4 Nano	Under-covered	76.5	1485	—	0 / 1 / 9	10 / 10	5	0 / 1	Repaired · 2026-07-01
E-786cc9019eb3ee2b	GPT-5.4 Nano	Under-covered	87.4	1499	—	0 / 0 / 10	10 / 10	5	0 / 1	Repaired · 2026-07-02
E-be95cc83ce87847c	GPT-5.4 Nano	Under-covered	78.1	1500	—	0 / 0 / 8	8 / 8	4	0 / 1	Repaired · 2026-07-02
E-51e935956fa76015	GPT-5.5	Under-covered	77.0	1499	—	0 / 0 / 8	8 / 8	4	0 / 1	First pass · 2026-07-02
E-ba4589e1cf2c827b	GPT-5.6 Luna	Under-covered	79.3	1501	—	0 / 0 / 8	8 / 8	4	0 / 1	First pass · 2026-07-10
E-0420ca9039edc6fb	GPT-5.6 Sol	Under-covered	93.2	1496	—	1 / 1 / 10	12 / 12	6	0 / 1	First pass · 2026-07-10
E-c6593bd789768bdb	GPT-5.6 Terra	Under-covered	88.3	1500	—	0 / 0 / 10	10 / 10	5	0 / 1	Repaired · 2026-07-10
E-fc495094a55d2f94	Ling 3.0 Flash	Under-covered	52.3	1486	—	0 / 1 / 5	6 / 6	3	0 / 1	First pass · 2026-07-24
E-a0c74818516a7f27	LongCat 2.0	Under-covered	41.7	1500	—	0 / 0 / 4	4 / 4	2	0 / 1	Repaired · 2026-07-24
E-1e46941293882535	Minimax M3	Under-covered	52.2	1485	—	0 / 1 / 5	6 / 6	3	0 / 1	First pass · 2026-07-02
E-d3d3da42e99695b1	Kimi K3	Pending	—	—	—	0 / 0 / 0	0 / 0	0	0 / 1	First pass · 2026-07-24
E-fc3ad6db6543001f	Kimi K3	Under-covered	78.1	1500	—	0 / 0 / 8	8 / 8	4	0 / 1	First pass · 2026-07-21
E-811f47e902ce0d21	Nex N2 Pro	Under-covered	88.3	1500	—	0 / 0 / 10	10 / 10	5	0 / 1	Repaired · 2026-07-11
E-e48b0ea99dfc517c	O3	Under-covered	78.1	1500	—	0 / 0 / 8	8 / 8	4	0 / 1	Repaired · 2026-07-21
E-18d1d6f3b7eabe0d	GPT-OSS 120B	Under-covered	78.1	1500	—	0 / 0 / 8	8 / 8	4	0 / 1	Repaired · 2026-07-21
E-d8c7a78cc7f666dc	Laguna S 2.1	Under-covered	65.0	1501	—	0 / 0 / 6	6 / 6	3	0 / 1	Repaired · 2026-07-24
E-5f64c2ae9860ba38	Qwen3.7 Max	Under-covered	88.8	1513	—	1 / 0 / 7	8 / 8	4	0 / 1	First pass · 2026-07-02
E-010dafcd7561851a	Qwen3.7 Plus	Under-covered	96.1	1500	—	0 / 0 / 12	12 / 12	6	0 / 1	Repaired · 2026-07-02
E-d910e47dd13c2c71	Hy3 Preview	Under-covered	78.1	1500	—	0 / 0 / 8	8 / 8	4	0 / 1	Repaired · 2026-07-02
E-66a4f0abd099d8f1	Inkling	Under-covered	75.4	1514	—	1 / 0 / 5	6 / 6	3	0 / 1	Repaired · 2026-07-24
E-1331dea6489316f2	Grok 4.5	Under-covered	78.1	1500	—	0 / 0 / 8	8 / 8	4	0 / 1	Repaired · 2026-07-21
E-1b0da3150a77b47b	Grok 4.5	Pending	—	—	—	0 / 0 / 0	0 / 0	0	0 / 1	Repaired · 2026-07-24
E-39b08fe4c0318552	MiMo-V2.5	Under-covered	77.9	1500	—	0 / 0 / 8	8 / 8	4	0 / 1	First pass · 2026-07-01
E-b9f44937a1269e8d	MiMo-V2.5-Pro	Under-covered	77.3	1499	—	0 / 0 / 8	8 / 8	4	0 / 1	First pass · 2026-07-02

Leaderboard

GB2-02: Chess leaderboard

Entrants are ranked by relative per-game score (0–100). Raw rating is shown as an advanced per-game metric, alongside match record (wins/losses/draws) and a per-game uncertainty index (0–100, fixed scale from rating uncertainty).

Reasoning level: Baseline Game: GB2-02: Chess

Scroll sideways to see every column.

GB2-02: Chess - Baseline filter
Rank	Model	Score	Contributing programs	Raw Elo	W / L / D	Uncertainty
1	DeepSeek V4 Pro	100.0	1	1514.6	1/0/9	100.0
2	Qwen3.7 Plus	96.1	1	1500.0	0/0/12	94.9
3	GPT-5.6 Sol	93.2	1	1496.4	1/1/10	94.9
4	DeepSeek V4 Flash	89.3	1	1501.3	0/0/10	100.0
5	Qwen3.7 Max	88.8	1	1513.3	1/0/7	100.0
6	Nex N2 Pro	88.3	1	1500.0	0/0/10	100.0
7	GPT-5.4 Mini	88.3	1	1500.0	0/0/10	100.0
8	GPT-5.6 Terra	88.3	1	1500.0	0/0/10	100.0
9	GPT-5.4 Nano	87.4	1	1498.8	0/0/10	100.0
10	GPT-5.4 Mini	86.7	1	1498.0	0/0/10	100.0
11	GPT-5.4 Nano	79.5	1	1488.9	0/1/9	100.0
12	GPT-5.6 Luna	79.3	1	1501.4	0/0/8	100.0
13	GPT-5.4	79.1	1	1501.2	0/0/8	100.0
14	GPT-5.4 Mini	79.0	1	1501.1	0/0/8	100.0
15	GPT-5.4 Nano	78.1	1	1500.0	0/0/8	100.0
16	DeepSeek V4 Flash	78.1	1	1500.0	0/0/8	100.0
17	Grok 4.5	78.1	1	1500.0	0/0/8	100.0
18	O3	78.1	1	1500.0	0/0/8	100.0
19	Kimi K3	78.1	1	1500.0	0/0/8	100.0
20	Hy3 Preview	78.1	1	1500.0	0/0/8	100.0
21	GPT-OSS 120B	78.1	1	1500.0	0/0/8	100.0
22	MiMo-V2.5	77.9	1	1499.8	0/0/8	100.0
23	MiMo-V2.5-Pro	77.3	1	1498.9	0/0/8	100.0
24	GPT-5.5	77.0	1	1498.6	0/0/8	100.0
25	GPT-5.4 Nano	76.5	1	1485.3	0/1/9	100.0
26	Inkling	75.4	1	1514.4	1/0/5	100.0
27	Laguna S 2.1	65.0	1	1501.4	0/0/6	100.0
28	Claude Opus 4.8	63.8	1	1500.0	0/0/6	100.0
29	Claude Opus 4.5	63.8	1	1500.0	0/0/6	100.0
30	Claude Opus 5	54.1	1	1515.6	1/0/3	100.0
31	Ling 3.0 Flash	52.3	1	1485.6	0/1/5	100.0
32	Minimax M3	52.2	1	1485.4	0/1/5	100.0
33	LongCat 2.0	41.7	1	1500.0	0/0/4	100.0
34	Claude Opus 5	0.0	1	1500.0	0/0/2	100.0